RU2515997C2 - Active queue management for wireless communication network uplink - Google Patents

Active queue management for wireless communication network uplink Download PDF

Info

Publication number
RU2515997C2
RU2515997C2 RU2011142315/07A RU2011142315A RU2515997C2 RU 2515997 C2 RU2515997 C2 RU 2515997C2 RU 2011142315/07 A RU2011142315/07 A RU 2011142315/07A RU 2011142315 A RU2011142315 A RU 2011142315A RU 2515997 C2 RU2515997 C2 RU 2515997C2
Authority
RU
Russia
Prior art keywords
base station
user equipment
transmission buffer
size
estimated
Prior art date
Application number
RU2011142315/07A
Other languages
Russian (ru)
Other versions
RU2011142315A (en
Inventor
Ифэн ТАНЬ
Риикка СУСИТАЙВАЛЬ
Йохан ТОРСНЕР
Original Assignee
Телефонактиеболагет Л М Эрикссон (Пабл)
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US16197209P priority Critical
Priority to US61/161,972 priority
Application filed by Телефонактиеболагет Л М Эрикссон (Пабл) filed Critical Телефонактиеболагет Л М Эрикссон (Пабл)
Priority to PCT/SE2009/050851 priority patent/WO2010107355A1/en
Publication of RU2011142315A publication Critical patent/RU2011142315A/en
Application granted granted Critical
Publication of RU2515997C2 publication Critical patent/RU2515997C2/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Application independent communication protocol aspects or techniques in packet data networks
    • H04L69/16Transmission control protocol/internet protocol [TCP/IP] or user datagram protocol [UDP]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • H04L47/12Congestion avoidance or recovery
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • H04L47/14Flow control or congestion control in wireless networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • H04L47/30Flow control or congestion control using information about buffer occupancy at either end or transit nodes
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • H04L47/31Tagging of packets, e.g. discard eligibility [DE] bit
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • H04L47/32Packet discarding or delaying
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Application independent communication protocol aspects or techniques in packet data networks
    • H04L69/16Transmission control protocol/internet protocol [TCP/IP] or user datagram protocol [UDP]
    • H04L69/163Adaptation of TCP data exchange control procedures
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W28/00Network traffic or resource management
    • H04W28/02Traffic management, e.g. flow control or congestion control
    • H04W28/10Flow control between communication endpoints
    • H04W28/12Flow control between communication endpoints using signalling between network elements
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W80/00Wireless network protocols or protocol adaptations to wireless operation
    • H04W80/06Transport layer protocols, e.g. TCP [Transport Control Protocol] over wireless

Abstract

FIELD: radio engineering, communication.
SUBSTANCE: invention relates to communication systems. According to the ideas presented herein, a base station implements active queue management (AQM) for uplink transmissions from user equipment (UE), e.g. a mobile terminal. The base station, e.g., an eNodeB in a long term evolution (LTE) network, uses buffer status reports, for example, to estimate packet delays for packets in the UE uplink transmit buffer. In one version of the method of AQM (base station) for the uplink includes estimating at least one of a transmit buffer size and a transmit buffer queuing delay for UE, and selectively dropping or congestion-marking packets received at the base station from the UE. The selective dropping or marking is based on the estimated transmit buffer size and/or the estimated transmit buffer queuing delay.
EFFECT: reduced signal transmission delay.
22 cl, 7 dwg

Description

FIELD OF TECHNOLOGY

The present invention generally relates to wireless networks, and in particular relates to active queue management (AQM) in such networks.

BACKGROUND

LTE is the new technology of the radio access network isolated from 3rd generation WCDMA, providing high peak bit rates and good end-to-end QoS. However, in many cases, the radio link is still a critical limitation of the end-to-end connection. In an overload situation, that is, a situation where the incoming data rate to the communication line is greater than the outgoing speed, excessive data is temporarily stored in the storage device. This storage device is often called a transfer buffer or queue. If congestion continues, the data queue will accumulate and become large. This can cause a number of problems, for example, large end-to-end delays, unfair separation between different threads, etc.

In addition, since the buffer is finite, the queue could ultimately exceed the physical limit, and some data would have to be discarded. An immediate way to solve this problem is to discard newly received data when the buffer is full. This approach is intuitive and easy to implement, but performance is far from optimal in terms of latency for the end user.

A more sophisticated approach for managing queues in a buffer is called Active Queue Management (AQM). AQM discards packets before the buffer is full. As a result, assuming that IP packets are being sent over the TCP / IP link, the TCP sender may sense a lost segment and, as a result, slow down the sending speed - see Stevens, W. TCP / IP Illustrated, Volume 1: The Protocols. Addison-Wesley, 1994. Thus, queue size and latency can be kept relatively low. Moreover, the throughput of the end-to-end communication line will not decrease significantly.

Significant work has been done for AQM. However, most AQM algorithms are designed for wired networks and are not suitable for mobile networks due to their changing bandwidth characteristics that result from changing radio conditions. On the other hand, the packet cancellation warning counter (PDPC) algorithm is one AQM algorithm for WCDMA - see Sågfors, M. Ludwig, R. Meyer, M. Peisa, J. Buffer Management for Rate-Varying 3G Wireless Links Supporting TCP Traffic . In the Proceedings of Vehicular Technology Conference, 2003, 3GPP also defined a simple algorithm called PDCP Reset in its specification - see 3GPP TS 36.321. Evolved Universal Terrestrial Radio Access (E-UTRA); Medium Access Control (MAC) protocol specification, version 8.2.0, May 2008. Another AQM algorithm is proposed for 3GPP in 3GGP R2-080937, On the Need for Active Queue Management for non-GBR Bearers, TSG-RAN WG2 Meeting # 60bis February 2008. This proposal was not accepted.

The PDCP Reset that is set for the UE in 3GPP Rel-8 is a simple delay-based algorithm that discards packets based on how long the packets have been in the PDCP queue. When the delay exceeds some predetermined threshold, packets will be discarded. This algorithm may support a small end-to-end delay, but in some situations can lead to a significant reduction in throughput.

The more sophisticated delay-based AQM algorithm proposed at the aforementioned TSG-RAN meeting implements a mechanism to increase PDCP discard throughput, for example by preventing consecutive packet discards without introducing much end-to-end delay. However, as noted, the proposal was not accepted in 3GPP for LTE Rel-8. You could specify a UE-based uplink AQM mechanism for LTE in Rel-9 (or later) to achieve better performance, but this would require standardization activities, and the mechanism would not be available for Rel-8 UEs.

US2008 / 186918A1 relates to lightweight active queue management. Queue management can be performed in a serving radio base station as well as in an access terminal, and an application that generates data packets can be executed locally or remotely for either the base station or the access terminal.

The 3GPP SDU Discard project, R2-074689, November 12, 2007, considers AQM as a mechanism that controls the size of the L2 data queue. AQM is defined as a function on the sender side, used to maintain the size of the queues on the sender side at an acceptable level, where the decision to drop the packet can be based, for example, on the size of the queue. It is proposed that the AQM mechanism should be defined for LTE, and that queuing should be modeled in PDCP, and the AQM mechanism should be located at the PDCP level.

The 3GPP project “Specifying SDU Discard function for the UE”, R2-074703, November 12, 2007 relates to the use of the SDU reset mechanism to implement the AQM mechanism as a function on the sender side for the LTE UE to control the size of the queue on the sender side and the corresponding queuing delays. AQM throws packets to force the upper layer protocols to reduce sending speed, thereby resulting in reduced queue size and reduced latency.

SUMMARY OF THE INVENTION

The main idea of the present invention is the implementation of an active queue management (AQM) mechanism in an uplink at a base station (eg, an eNodeB) instead of a UE.

A related problem is that in the aforementioned delay-based AQM algorithms, the delay of each packet in the queue must be known. UE buffer delays are mainly known only on the UE side, and there is no standardized way to report their eNodeBs. The ideas proposed in this document allow the eNodeB or other base station to obtain information about the delays of the UE buffers.

In relation to LTE, the AQM algorithm for the uplink LTE is implemented in the eNodeB, which provides a number of advantages, including: the network has full control over the configuration of the algorithm; no standardization activity is needed (the solution is patented); and the mechanism also works for Rel-8 UEs (unlike a given UE solution for Rel-9 or later).

Thus, in accordance with the ideas presented for one or more embodiments in this document, the network base station estimates the queue delay in the transmission buffer in the UE based on the received buffer status reports from the UE and the amount of data served. The AQM algorithm in a base station, such as an eNodeB, uses the estimated delay and size of the UE buffer obtained from the status reports of the UE buffer to decide when packets should be discarded in the base station to control the UE buffer. As a result, TCP will slow down in response to a packet drop, and the queue size will remain small.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a printout of pseudo code for one embodiment of an active queue management method that is discussed in this document.

FIG. 2 is a block diagram of one embodiment of a base station configured to provide active queue management for a user equipment (UE).

FIG. 3 is a printout of pseudo code for one embodiment of a delay estimation method for active queue management.

FIG. 4 is a block diagram of one embodiment of a wireless communication network, such as an LTE network, in which an included base station provides active queue management for one or more mobile terminals or other user equipment.

FIG. 5 is a block diagram of one embodiment of a base station, such as an eNodeB, that is configured with one or more processors that implement active queue management for user equipment.

FIG. 6 is a flowchart of one embodiment of a method in which a base station provides active queue management for user equipment.

FIG. 7 is a flowchart of another embodiment of a method in which a base station provides active queue management for user equipment.

DETAILED DESCRIPTION OF SOME EMBODIMENTS

Compared to a queue size based approach (e.g., PDPC), active queue management (AQM) can support both greater link utilization and less end-to-end latency. In addition, bandwidth fluctuations are less affected by AQM. Thus, in the new algorithm proposed in the present invention, the delay-based principle is retained. However, it is preferable that a base station, for example an eNodeB in an LTE network, provides delay-based queue management for a user equipment (UE), for example, a mobile terminal.

To apply the principle in an eNodeB or other base station in a wireless communication network, delays in the packet queue in the buffer of the UE must be known by the eNodeB. However, the delay of the UE queue is not known directly in the eNodeB and should be estimated. A general description of the proposed AQM algorithm and the providing method for estimating the delay for one or more embodiments of the present invention are disclosed below.

Delay based AQM in eNodeB

In LTE, buffer status reports (BSRs) are transmitted from the UE to inform the eNodeB of the size of the queue in the buffer of the UE. BSRs are triggered by special events defined by 3GPP - see 3GPP TS 36.321. Evolved Universal Terrestrial Radio Access (E-UTRA); Medium Access Control (MAC) protocol specification, version 8.2.0, May 2008. Typically, the initiation interval is approximately tens of milliseconds. Thus, BSR can be understood as a monitor of the queue length with a short time interval.

Assuming that packet delay can be estimated on the eNodeB side based on the BSR and the amount of data served, the delay-based AQM algorithm can be implemented in the eNodeB using the estimated packet delay. FIG. 1 illustrates pseudo-code for an algorithm in one embodiment of a proposed delay-based AQM method in an eNodeB. lowerDropThreshold, minAgeThreshold and minInterDropTime are three configurable parameters. Each time an eNodeB receives a BSR element, it first compares the queue size reported by the BSR with lowerDropThreshold; if the size reported by the BSR is less than the lowerDropThreshold, a reset is not allowed. Otherwise, the delay will be estimated using the method described later. If the estimated delay is greater than minAgeThreshold, the packet will be discarded until the time interval between the current time and the previous packet reset time is less than minInterDropTime.

One implementation of the new AQM algorithm, which is discussed in this document, is illustrated in FIG. 2. Packets are discarded at the Radio Control (RLC) level when the RLC SDUs are delivered to higher layers in the eNodeB. Implementation can also be done at the PDCP level, in which case the discarded data units would be PDCP SDUs. Dropping data units at the lower levels is not possible because the RLC protocol in the eNodeB would request retransmission of the dropped packet. As a result, the TCP sender would not be able to detect the loss of a segment, and accordingly would not reduce the sending speed.

You can also use information other than the BSR to estimate the size of the UE buffer in the eNodeB. For example, 3GPP discussed the introduction of bits in the MAC header, which carry information to help the eNodeB scheduler (commonly known as "happy bits"). These bits could, for example, contain a short buffer status report or queue delay information for packets in the UE. If such bits are introduced in a later version of the 3GPP standard, then you can use them instead or in combination with the BSR to estimate the size of the UE buffer in the eNodeB.

It should be noted that, as an alternative to dropping packets in the AQM algorithm, it is also possible to “mark” packets by setting the so-called ECN bits in the IP header (ECN = explicit congestion notification). If ECN bits are set, then TCP will also slow down as a result. Thus, the approaches to AQM described in this document can be adapted to discard or label packets.

Detailed Description of One Embodiment of a Method for Estimating Queue Delay

In this subsection, an embodiment of an AQM method is provided for estimating a queue delay of head data, that is, the oldest data, in an arbitrary buffer. It is assumed that the queue length can be controlled discretely, that is, you can know the queue length at discrete time intervals. It is also assumed that the amount of data serviced by the buffer during each time interval is available (either from the amount of data scheduled for transmission or from the amount of data received in the eNodeB). Also, the length of the observation time interval is known and assumed to be small.

For convenience, the following text uses t i to indicate the time when the queue is being monitored, while the length of the queue in t i is denoted by Q i . Moreover, the time interval between t i − 1 and t i is denoted by Δt i , and data served during the time interval Δt i is denoted by L i . Finally, let R i denote the amount of data received in the queue during the time interval. The new data R i is not directly known, but can be calculated from the length of the queue and outgoing bits, i.e.

R i = Q i + L i -Q i-1 ,

where Q i , L i , Q i-1 are known variables.

At time n, the queue delay for the oldest data in the queue, assuming that nothing is served during the observation interval, is

Figure 00000001

where r is the index of the observation event t r when the leading data arrived, that is, the oldest data that still exists in the queue. However, during the observed time interval, data corresponding to L n is already removed from the queue. Thus, the actual head delay in the head data depends on the region in which Q n is located. For example, suppose that Q n lies between

Figure 00000002
, then the maximum queue delay is calculated as the average of the corresponding delays, i.e.

Figure 00000003

The pseudocode of a detailed embodiment of a method for estimating a delay (packet) is shown in FIG. 3. In the algorithm, the number of bits that are not yet served is stored along with the corresponding delays. Values are stored in data structures indicated by R [] and D []. When the current queue length, the amount of data served by the buffer, L i , and the time interval Δt i are received, then based on these values, the values of the served bits are removed from R [], and the delays of all bits are updated.

Additional Exemplary Embodiments

FIG. 4 illustrates a UE 10, for example, a cell phone or other mobile terminal having packet data connection capabilities. The UE 10 has an information connection 12 with a communication node 14, which may be another UE or essentially any communication device, server, or other system capable of establishing an end-to-end information connection with the UE 10. In the illustration, the information connection 12 may be a TCP / IP communication line.

More specifically, a wireless communication network 20, such as an LTE network, supports an information connection 12 between the UE 10 and the communication node 14 by communication connecting the UE 10 to one or more external data networks 22, such as the Internet. Of course, the same type of data connection can be supported by the network 20 for UEs operating in the network 20, and active queue management at base stations, with respect to the UE, can be performed for data connections established inside and outside the network 20.

In more detail, the network 20 includes a Radio Access Network (RAN) 24, which is communicatively connected to a Core Network 26 (CN), which in turn is communicatively connected to an external network (s) 22. RAN 24 includes a number of base stations 28 where for simplicity one is shown. Base station 28, which may be an eNodeB in an embodiment of an LTE network 20, advantageously includes one or more processing schemes that are configured to provide active queue management for transmission buffers in any number of UE 10 having data connections 12 established through the base station 28.

Thus, using a TCP / IP communication link established between the UE 10 and the communication node 14 through the base station 28, the base station 28 can initiate TCP based flow control for uplink transmissions from the UE 10 based on selectively dropping or marking upstream packets (IP packets) sent from the UE 10 to the communication node 14. That is, the loss (or marking during congestion) of IP packets on the uplink will cause TCP to reduce the sending speed of UE packets and thereby reduce congestion.

FIG. 5 illustrates a non-limiting example embodiment of a base station 28. The illustrated base station 28 includes transceiver circuits 30 for signaling a downlink and an uplink to and from a potentially large number of UEs 10. Base station 28 further includes CN interface circuits 32 for communicating with CN 26. Furthermore, base station 28 includes a number of communication and processing control circuits 34, which may include a set of rack / basket processing cards or other subsystems processing.

Regardless of the particular physical implementation, base station 28 further includes an active queue management (AQM) processor 40 for implementing active queue management in transmission buffers in one or more UEs 10. (The circuit may be configured to provide AQMs for multiple UEs 10, or parallel implementations of the circuit may be used to provide AQM for several UEs 10.) Regarding such flexibility, it will be appreciated that the processing circuits 34 in one or more embodiments include one or more microprocessors and / or circuits of a digital signal processor together with associated memory for programs and data. As such, the illustrated circuits may represent implementations of functional circuits at least partially implemented by executing computer program instructions stored on a computer-readable medium in base station 28.

In an exemplary implementation of the AQM processor 40, the discard / tag controller selectively determines whether to discard (or mark) packets from the UE based on the estimation of at least one of the estimated transmission buffer size for the UE 10 and the estimated queue delay for packets in the transmission buffer of the UE . Supporting such functionality, the AQM processor 40 includes or is associated with one or more timers, a transmission buffer size estimator, and a queue delay estimator, and it will be necessary to take into account that the working memory of data and programs may be included or associated with these circuits .

For further reference, an exemplary embodiment of UE 10 is shown including transceiver circuits 50, transmit processor 52, and transmit buffer 54. Then, for the illustrated example, the AQM processor 40 in the base station 28 will be perceived accordingly as providing AQM for the transmission buffer 54 in the UE 10.

With the support of the functional processing shown in FIG. 5, it will be appreciated that one embodiment of the method disclosed herein comprises a method at a base station for controlling a transmission queue in user equipment transmitting packets of a base station. As shown in the exemplary embodiment of FIG. 6, the method includes estimating at least one of a transmission buffer size and queue delay in a transmission buffer for user equipment (step 100), and selectively discarding or marking when congesting packets received at the base station from the user equipment based on at least one of the estimated transmission buffer size and the estimated queue delay in the transmission buffer (step 102).

Packets in one or more embodiments are Internet Protocol (IP) packets sent by user equipment over a Transmission Control Protocol (TCP) / IP information connection. Essentially, selective dropping or marking when congesting packets received at the base station from the user equipment includes selective dropping or marking when congesting IP packets to initiate TCP-based flow control in the data connection.

In one or more embodiments, selective drop or marking on packet congestion comprises selectively dropping packets at the Radio Control (RLC) level or at the Packet Data Convergence Protocol (PDCP) level to avoid initiating retransmission of packets based on RLC or PDCP dropped packets by the user equipment .

Additionally, as detailed earlier in this document, selective discarding or marking when congesting packets received at the base station from the user equipment includes selective discarding or marking when congesting packets received from the user equipment if the estimated size of the transmission buffer exceeds a predetermined size threshold buffers. In one or more embodiments, estimating a queue size in a user equipment transmission buffer is based on receiving buffer status reports or buffer status indicators from the user equipment. Similarly, estimating the queue delay in a transmission buffer can also be based on buffer status reports or buffer status indicators, and the amount of data served, which is calculated from the amount of uplink data transmitted or scheduled for transmission from the user equipment to the base station.

In at least one embodiment, estimating a queue delay in a transmission buffer comprises calculating the amount of data not yet served from the user equipment transmission buffer based on buffer status reports or buffer status indicators, and the amount of serviced or scheduled data, and storing and updating this amount of data together with a corresponding delay in the data structure supported at the base station. Here, the estimated queue delay in the transmission buffer can be calculated as the delay of the oldest record in the data structure.

In another embodiment, selectively discarding or marking when congesting packets received at the base station from the user equipment, comprising selectively discarding or marking when congesting packets received from the user equipment, if the estimated transmission buffer size exceeds a predetermined threshold size of the buffer and if the estimated queue delay in the transmission buffer exceeds a predetermined threshold value of the queue delay. More specifically, the method may not include discarding or marking during packet congestion if the estimated size of the transmission buffer is below a predetermined size threshold.

Based on this approach, the method may further include selectively discarding or marking when congesting packets received at the base station from user equipment, further based on checking whether the estimated queue delay in the transmission buffer exceeds a predetermined delay threshold value if the estimated transmission buffer size above a given size threshold. If this is the case, then the method discards or marks during overloading the currently received packet, while the time interval from the last discarding or marking event of the packet is not less than the specified threshold time value.

FIG. 7 illustrates an embodiment of the method described immediately above. In accordance with the illustrated processing logic that may be implemented for a given packet using the AQM processor 40 of FIG. 5, the estimated transmission buffer size of the UE designated as EBS is compared with a predetermined threshold size of the buffer size designated as TH (S) (step 110). If the estimated size of the transmission buffer is below a predetermined size threshold (step 112, EBS does not exceed TH (S)), then the specified packet is not discarded or otherwise marked (step 114).

However, if the threshold size of the buffer size is exceeded, processing continues by comparing the estimated queue delay in the transmission buffer of the UE designated as EQD with the predetermined threshold value of the delay designated as TH (D) (steps 116 and 118). If the estimated queue delay does not exceed a predetermined delay threshold, then the predetermined packet is not discarded / marked (step 120). However, if the threshold value of the queue delay is exceeded, then the specified packet can be discarded / marked at this moment, or additional verification is performed in at least one embodiment.

As shown in the figure, an additional check may be to determine if the time interval since the last discard / mark event has exceeded a predetermined interval timer (INT.> TMR?) (Steps 122 and 124). If less than the specified timer duration has elapsed since the last reset / marking event, then the specified packet is not discarded or otherwise marked (step 126). If the minimum elapsed time has passed, then discarding (or marking) is performed (step 128).

This time spent check prevents discarding or other marking when overloading too many packets and prevents discarding / marking of consecutive packets. Of course, the timer used to post drop / mark events can be updated or changed dynamically based on performance monitoring or another metric associated with the data channel, like other predetermined threshold values used for AQM.

Non-limiting Advantages of the Invention

The solution proposed by the present invention has the following characteristics, but is not limited to: (1) it is implemented in the network (for example, at the base station), and not at the UE, providing accordingly better control of the UE buffer from the network side; (2) no standardization activity is required (patented solution); (3) the solution also works for existing Rel-8 UEs, which would not be true for a standardized UE solution in Rel-9 or later; (4) It supports low end-to-end packet delay without significant impact on throughput.

In one or more embodiments, the present invention comprises an uplink mechanism for selectively discarding (or marking) received packets at a base station to improve TCP performance. In at least one embodiment, the discarding (or marking) criteria is based on the estimated size of the UE buffer and, if desired, on the estimated delay in the UE buffer.

The size of the UE buffer is estimated, for example, from the received status reports of the UE buffer, and the delay in the UE buffer is estimated, for example, based on the received buffer reports and the amount of data served. In support of this estimate, the amount of data served is computed from the amount of scheduled uplink data or the amount of received data at the base station (eNodeB).

Abbreviations

The following abbreviations are used:

AQM: Active Queue Management;

ARQ: Automatic repeat request;

BSR: Buffer Status Report;

eNodeB: Enhanced Node B;

LTE: Long-Term Development Project;

PDCP: Packet Data Convergence Protocol;

PDPC: packet transmission cancel warning counter;

PDU: Protocol data unit;

QoS: Quality of Service;

RLC: Radio Control; and

SDU: Service Data Unit.

Those skilled in the art will appreciate that the present invention is not limited to the above examples and illustrations, but rather is limited only by the following appended claims and their legal equivalents.

Claims (22)

1. A method of selectively discarding or marking during congestion, at a base station, packets received from user equipment (10), characterized in that the active control of the uplink queue in the transmission buffer (54) in the user equipment is carried out in the said base station (28 ) through the steps in which:
evaluating (100) in said base station (28) at least one of a transmission buffer size and a queue delay in a transmission buffer for a transmission buffer in user equipment (10); and
selectively discarding or marking at congestion (102) in said base station (28) packets received at base station (28) from user equipment (10) based on at least one of the estimated transmission buffer size and the estimated queue delay in transfer buffer.
2. The method according to claim 1, in which the packets are Internet Protocol (IP) packets received from the user equipment (10) over the information connection of the Transmission Control Protocol (TCP) / IP, and in which the step of selectively discarding or marking when the packets are overloaded received at the base station (28) from the user equipment (10) comprises selectively dropping or marking when congesting IP packets to initiate TCP-based flow control in the data connection.
3. The method according to claim 1, wherein the step of selectively dropping or marking when congesting packets comprises selectively dropping packets at a Radio Control (RLC) level or at a Packet Data Convergence Protocol (PDCP) level to avoid initiating RLC-based retransmission of packets or PDCP discarded packets by user equipment (10).
4. The method according to claim 1, wherein the step of selectively dropping or marking when congesting packets received at the base station (28) from the user equipment (10), comprises selectively dropping or marking when congesting packets received from the user equipment (10), if the estimated size of the transmission buffer exceeds a predetermined threshold size of the buffer size.
5. The method according to claim 1, wherein the step of selectively discarding or marking when congesting packets received at the base station (28) from the user equipment (10), comprises selectively discarding or marking when congesting packets received from the user equipment (10), if the estimated size of the transmission buffer exceeds a predetermined threshold size of the buffer and if the estimated delay of the queue in the transmission buffer exceeds a predetermined threshold value of the delay of the queue.
6. The method according to claim 1, further comprising a step at which the size of the queue in the transmission buffer of the user equipment is estimated at the base station (28) based on the receipt of buffer status reports or buffer status indicators from the user equipment (10).
7. The method according to claim 6, further comprising a step at which the queue delay in the transmission buffer is estimated at the base station (28) based on buffer status reports or buffer status indicators and the amount of data served, which is calculated from the amount of transmitted or scheduled for transmission uplink data from user equipment (10) to base station (28).
8. The method according to claim 1, wherein the step of selectively dropping or marking when congesting packets received at the base station (28) from user equipment (10) includes not dropping or not marking when congesting packets, if the estimated size of the transmission buffer below a given size threshold.
9. The method according to claim 8, in which the step of selectively discarding or marking when congesting packets received at the base station (28) from the user equipment (10), if the estimated size of the transmission buffer is higher than a predetermined threshold size, further comprises checking if estimated delay of the queue in the transmission buffer, the specified delay threshold value, and if so, then discard or mark upon congestion the currently received packet, until the time interval from the last packet drop or marking event e is less than a predetermined threshold time.
10. The method according to claim 1, further comprising a step at which the queue delay in the transmission buffer is estimated at the base station (28) by calculating the amount of data not yet served from the transmission buffer (54) of the user equipment (10) based on the buffer status reports or indicators of the status of the buffer and the amount of serviced or scheduled data and save and update this amount of data along with the corresponding delay in the data structure supported at the base station (28).
11. The method of claim 10, in which the estimated queue delay in the transmission buffer is calculated as the delay of the oldest record in the data structure.
12. The base station (28), which includes one or more processing schemes (40), configured to selectively drop or mark when reloading packets received from user equipment (10) to ensure active queue management in the transmission buffer (54) in user equipment (10) based on:
estimating at least one of a transmission buffer size and queue delay in a transmission buffer for a transmission buffer in user equipment (10); and
selectively discarding or marking when congesting packets received at the base station (28) from user equipment (10) based on at least one of the estimated transmission buffer size and the estimated queue delay in the transmission buffer.
13. The base station of claim 12, wherein the packets are Internet Protocol (IP) packets received from user equipment over the Transmission Control Protocol (TCP) / IP information connection, and the base station (28) is configured to selectively drop or mark when congestion of packets received at the base station (28) from user equipment (10) by selectively dropping or marking when congesting IP packets to initiate TCP-based flow control in the data connection.
14. The base station of claim 12, which is configured to selectively discard or mark when congesting packets by selectively discarding packets at the Radio Control (RLC) level or at the Packet Data Convergence Protocol (PDCP) level to avoid initiating RLC-based packet retransmission or PDCP discarded packets by user equipment (10).
15. The base station according to claim 12, which is configured to selectively drop or mark when reloading packets received at the base station (28) from the user equipment (10) by selectively drop or mark when reloading packets received from the user equipment (10) if the estimated transmission buffer size exceeds a predetermined threshold size of the buffer size.
16. The base station according to claim 12, which is configured to selectively drop or mark when reloading packets received at the base station (28) from user equipment (10) by selectively drop or mark when reloading packets received from user equipment (10) if the estimated transmission buffer size exceeds a predetermined buffer size threshold and if the estimated queue delay in the transmission buffer exceeds a predetermined queue delay threshold.
17. The base station according to claim 12, which is configured to estimate the size of the queue in the transmission buffer of the user equipment based on the receipt of buffer status reports or buffer status indicators from the user equipment (10).
18. The base station according to claim 17, which is configured to estimate queue delay in the transmission buffer based on buffer status reports or buffer status indicators and the amount of data served, which is calculated from the amount of uplink data transmitted or scheduled for transmission from user equipment ( 10) to the base station (28).
19. The base station according to claim 12, which is configured to selectively discard or mark when congesting packets received at the base station (28) from user equipment (10) by rejecting discarding or marking when congesting packets, if the estimated size of the transmission buffer is lower given threshold size.
20. The base station according to claim 19, which, if the estimated size of the transmission buffer is higher than a predetermined size threshold value, is configured to selectively discard or mark when congesting packets received at the base station (28) from user equipment (10) by further checking that whether the estimated queue delay in the transmission buffer exceeds the specified delay threshold, and if so, discarding or marking when the current received packet is overloaded, until the time interval from the last syvaniya or marking the packet is not less than a predetermined threshold time.
21. The base station according to claim 12, which is configured to estimate the queue delay in the transmission buffer by calculating the amount of data not yet served from the user equipment transfer buffer (54) (10) based on buffer status reports or buffer status indicators and the amount of serviced or planned data and storing and updating this amount of data along with a corresponding delay in the data structure supported at the base station (28).
22. The base station according to item 21, which is configured to calculate the estimated queue delay in the transmission buffer as a delay of the oldest record in the data structure.
RU2011142315/07A 2009-03-20 2009-07-01 Active queue management for wireless communication network uplink RU2515997C2 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US16197209P true 2009-03-20 2009-03-20
US61/161,972 2009-03-20
PCT/SE2009/050851 WO2010107355A1 (en) 2009-03-20 2009-07-01 Active queue management for wireless communication network uplink

Publications (2)

Publication Number Publication Date
RU2011142315A RU2011142315A (en) 2013-04-27
RU2515997C2 true RU2515997C2 (en) 2014-05-20

Family

ID=41566273

Family Applications (1)

Application Number Title Priority Date Filing Date
RU2011142315/07A RU2515997C2 (en) 2009-03-20 2009-07-01 Active queue management for wireless communication network uplink

Country Status (5)

Country Link
US (1) US8964539B2 (en)
EP (1) EP2409516B1 (en)
CN (1) CN102742320B (en)
RU (1) RU2515997C2 (en)
WO (1) WO2010107355A1 (en)

Families Citing this family (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110282980A1 (en) * 2010-05-11 2011-11-17 Udaya Kumar Dynamic protection of a resource during sudden surges in traffic
US20120163205A1 (en) * 2010-06-28 2012-06-28 Qualcomm Incorporated System and method for flow control in a multi-point hsdpa communication network
US8989140B2 (en) 2010-06-28 2015-03-24 Qualcomm Incorporated System and method for mobility in a multi-point HSDPA communication network
US8891356B2 (en) 2010-06-28 2014-11-18 Qualcomm Incorporated System and method for multi-point HSDPA communication utilizing a multi-link RLC sublayer
US8989004B2 (en) 2010-11-08 2015-03-24 Qualcomm Incorporated System and method for multi-point HSDPA communication utilizing a multi-link PDCP sublayer
EP2679047B1 (en) 2011-02-22 2014-09-17 Telefonaktiebolaget L M Ericsson (PUBL) Method and device for congestion situations
US8737211B2 (en) 2011-08-03 2014-05-27 Qualcomm Incorporated Methods and apparatuses for network configuration of user equipment communication modes in multiflow systems
US9125098B2 (en) 2011-08-03 2015-09-01 Qualcomm Incorporated Method and apparatus for flow congestion control in multiflow networks
US9386128B2 (en) * 2012-03-23 2016-07-05 Qualcomm Incorporated Delay based active queue management for uplink traffic in user equipment
US9473974B2 (en) 2012-05-30 2016-10-18 The University Of Hong Kong Enhancing AQM to combat wireless losses
WO2014003327A1 (en) 2012-06-26 2014-01-03 Lg Electronics Inc. Method and apparatus for cancelling buffer status report in wireless communication system
EP2685757B1 (en) * 2012-07-10 2014-12-17 Telefonaktiebolaget L M Ericsson (PUBL) Technique for handling congestion control
US8885476B2 (en) * 2012-10-04 2014-11-11 Verizon Patent And Licensing Inc. TCP flow control optimized for networks having radio segments
US9286620B2 (en) * 2012-11-05 2016-03-15 Broadcom Corporation Annotated tracing for data networks
CN102984083B (en) * 2012-11-19 2018-07-24 南京中兴新软件有限责任公司 queue management method and device
US9538435B1 (en) * 2012-11-20 2017-01-03 Sprint Communications Company L.P. Volte packet delay based network configuration
US10136355B2 (en) 2012-11-26 2018-11-20 Vasona Networks, Inc. Reducing signaling load on a mobile network
CN103905143B (en) * 2012-12-27 2017-05-24 重庆重邮信科通信技术有限公司 Uplink-data processing method and system
US9106557B2 (en) * 2013-03-13 2015-08-11 Comcast Cable Communications, Llc Scheduled transmission of data
US9106391B2 (en) * 2013-05-28 2015-08-11 International Business Machines Corporation Elastic auto-parallelization for stream processing applications based on a measured throughput and congestion
EP3005601B1 (en) * 2013-05-31 2019-04-03 Telefonaktiebolaget LM Ericsson (publ) Predictive scheduling for uplink transmission in a cellular network
US9148814B2 (en) 2013-10-28 2015-09-29 At&T Intellectual Property I, L.P. Probe mechanism for discovering explicit congestion notification data
US10341881B2 (en) 2013-11-12 2019-07-02 Vasona Networks, Inc. Supervision of data in a wireless network
US10039028B2 (en) 2013-11-12 2018-07-31 Vasona Networks Inc. Congestion in a wireless network
US9345041B2 (en) * 2013-11-12 2016-05-17 Vasona Networks Inc. Adjusting delaying of arrival of data at a base station
US9397915B2 (en) 2013-11-12 2016-07-19 Vasona Networks Inc. Reducing time period of data travel in a wireless network
KR20160028792A (en) 2014-09-04 2016-03-14 삼성전자주식회사 Apparatus and method for rate control in wireless communication system
CN105636098A (en) * 2014-10-30 2016-06-01 中兴通讯股份有限公司 Method and device for estimating waiting time delay of uplink business
EP3035615A1 (en) * 2014-12-17 2016-06-22 Alcatel Lucent Method for queue management
KR20160137237A (en) 2015-05-22 2016-11-30 삼성전자주식회사 Method and apparatus for buffer management in wireless communication system
US20190007291A1 (en) * 2015-08-11 2019-01-03 Lg Electronics Inc. Method for performing uplink packet delay measurements in a wireless communication system and a device therefor
US10412634B2 (en) 2015-08-13 2019-09-10 Telefonaktiebolaget Lm Ericsson (Publ) Predictive adaptive queue management
US20180352460A1 (en) * 2015-09-24 2018-12-06 Nokia Solutions And Networks Oy Reporting of Uplink (UL) Quality of Service (QoS) Metrics
US9756654B2 (en) 2015-10-28 2017-09-05 Cisco Technology, Inc. Timer-based scheme for user equipment queue state estimation

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002082747A2 (en) * 2001-04-09 2002-10-17 Telefonaktiebolaget L M Ericsson (Publ) Method of controlling a queue buffer
RU2329607C2 (en) * 2003-12-23 2008-07-20 Телефонактиеболагет Лм Эрикссон (Пабл) Method and device to control queue buffer

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6901593B2 (en) * 2001-05-08 2005-05-31 Nortel Networks Limited Active queue management with flow proportional buffering
GB2372172B (en) * 2001-05-31 2002-12-24 Ericsson Telefon Ab L M Congestion handling in a packet data network
AUPR688201A0 (en) * 2001-08-09 2001-08-30 University Of Melbourne, The An active queue management process
KR101000391B1 (en) * 2003-09-01 2010-12-13 엘지전자 주식회사 method of controlling data rate over reverse link
US7706403B2 (en) 2003-11-25 2010-04-27 Telefonaktiebolaget Lm Ericsson (Publ) Queuing delay based rate control
JP2007524262A (en) 2003-12-23 2007-08-23 テレフオンアクチーボラゲット エル エム エリクソン(パブル) Method and apparatus for controlling queue buffer
US7701852B1 (en) * 2004-08-26 2010-04-20 Sprint Communications Company L.P. Method for analyzing performance of a network through measuring and reporting delay in routing devices
WO2007024120A1 (en) * 2005-08-26 2007-03-01 Electronics And Telecommunications Research Institute Method for requesting resource and scheduling for uplink traffic in mobile communication and apparatus thereof
US9119220B2 (en) * 2005-12-22 2015-08-25 Qualcomm Incorporated Methods and apparatus for communicating backlog related information
CN100596099C (en) * 2006-06-23 2010-03-24 华为技术有限公司 Radio access bear data transmission method
US8630256B2 (en) 2006-12-05 2014-01-14 Qualcomm Incorporated Method and system for reducing backhaul utilization during base station handoff in wireless networks
RU2432698C2 (en) * 2007-03-14 2011-10-27 Интердиджитал Текнолоджи Корпорейшн Method and device for preventing uplink attenuation in long-term development system
EP2028890B1 (en) * 2007-08-12 2019-01-02 LG Electronics Inc. Handover method with link failure recovery, wireless device and base station for implementing such method
TWI376112B (en) * 2007-10-29 2012-11-01
JP4952586B2 (en) * 2008-01-07 2012-06-13 富士通株式会社 Packet data discarding method, radio communication apparatus, and mobile communication system
US8040802B2 (en) * 2008-01-14 2011-10-18 Alcatel Lucent Method of controlling congestion for wireless communications
US8248941B2 (en) * 2008-02-01 2012-08-21 Nokia Siemens Networks Oy Method, apparatus and computer program for uplink scheduling in a network that employs relay nodes
US9584416B2 (en) * 2009-06-08 2017-02-28 Qualcomm Incorporated Systems and methods to provide flow control for mobile devices
US8923115B2 (en) * 2009-08-25 2014-12-30 Telefonaktiebolaget L M Ericsson (Publ) Using the ECN mechanism to signal congestion directly to the base station

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002082747A2 (en) * 2001-04-09 2002-10-17 Telefonaktiebolaget L M Ericsson (Publ) Method of controlling a queue buffer
RU2329607C2 (en) * 2003-12-23 2008-07-20 Телефонактиеболагет Лм Эрикссон (Пабл) Method and device to control queue buffer

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
3GPP TSG-RAN WG2 #60, Ericsson, SDU Discard, R2-074689, 11.2007, http://www.3gpp.org/ftp/tsg_ran/WG2_RL2/TSGR2_60/Docs/R2-074689.zip. 3GPP TSG-RAN WG2 #60, Ericsson, Specifying SDU Discard function for the UE, R2-074703, 11.2007, http://www.3gpp.org/ftp/tsg_ran/WG2_RL2/TSGR2_60/Docs/R2-074703.zip *

Also Published As

Publication number Publication date
CN102742320B (en) 2016-04-06
US8964539B2 (en) 2015-02-24
US20120039169A1 (en) 2012-02-16
WO2010107355A1 (en) 2010-09-23
CN102742320A (en) 2012-10-17
EP2409516B1 (en) 2013-03-13
RU2011142315A (en) 2013-04-27
EP2409516A1 (en) 2012-01-25

Similar Documents

Publication Publication Date Title
EP1516501B1 (en) Method and apparatus for enhancing the quality of service of a wireless communication
CA2677112C (en) Method and apparatus for enhancing rlc for flexible rlc pdu size
JP4988804B2 (en) Cognitive flow control based on channel quality conditions
EP2634950B1 (en) Method and Apparatus for Power Headroom Reporting
US7489691B2 (en) Scheduling retransmission in access networks
EP2749118B1 (en) Improving tcp performance in a cellular network
TWI483642B (en) Method and apparatus for reporting a buffer status
EP2134132B1 (en) Method and apparatus for handling scheduling information report
JP5166420B2 (en) Method and apparatus for resource management in wireless devices
EP1836866B1 (en) Uplink congestion detection and control between nodes in a radio access network
US8885542B2 (en) Quality of service control in a relay
CN100499858C (en) Radio communication network and flow control method
US20050259662A1 (en) Method and apparatus for scheduling enhanced uplink dedicated channels in a mobile communication system
US20070015525A1 (en) Coordinated data flow control and buffer sharing in umts
US20060203760A1 (en) Base station device and transmission method
ES2315369T3 (en) Treatment of congestion and delay in a data network by packages.
JP5116845B2 (en) Jitter-based media layer adaptation in real-time communication systems
CN101849391B (en) Efficient flow control in a radio network controller (RNC)
EP2479944A1 (en) Method and apparatus for supporting uplink starvation avoidance in a long term evolution system
US8923115B2 (en) Using the ECN mechanism to signal congestion directly to the base station
EP2109954B1 (en) Ack prioritization in wireless networks
ES2376060T3 (en) Congestion control within a radio access network.
KR20050014705A (en) Method and apparatus for scheduling transmissions in wireless data networks
US20110222406A1 (en) Method And Device For Enabling Indication Of Congestion In A Telecommunications Network
US9112797B2 (en) Congestion handling in a communication network